Quantitative Finance · Wed, Nov 4, 2026

arXiv Quantitative Finance (q-fin).

RSS feed
arXiv q-fin.PM

Mitigating Return Drag and Regime Whipsaws in Sector Rotation: An Asymmetric Downside-Aware Nested Clustered Optimization Framework

In briefReducing Return Drag and Regime Switching in Sector Rotation: An Asymmetric Downside-Risk Nested Clustered Optimization Framework

SummaryThis paper addresses two weaknesses in sector rotation strategies: symmetric variance penalizes upside and downside equally, causing return drag, and Hidden Markov Model regimes switch too often, raising turnover. The authors propose Asym-Sticky-NCO, which combines a Bawa-Lindenberg downside semi-covariance matrix with a Gaussian HMM that has a persistence prior, applied to all 11 S&P 500 GICS sectors. In 65 out-of-sample months from August 2020 to December 2025, the framework reports a 13.41% CAGR and a 14.22% maximum drawdown, with regime switches cut from 16 to 4 and monthly turnover reduced from 60.78% to 30.48%.

Supriadi

arXiv q-fin.TR · q-fin.ST

Beyond Anonymous Order Flow: Wallet Identity as a Feature for Short-Horizon Return Prediction

In briefWallet Identity as a Predictive Feature in Hyperliquid Order Flow

SummaryThis paper uses Hyperliquid's Level-4 data to reconstruct full-depth order books from consensus-ordered messages and tests whether pseudonymous wallet identity helps predict short-horizon returns. The July 1–27, 2026 dataset covers 17.1 billion messages, 14.3 million aggressive orders from 147,113 wallets, and $84.3 billion in taker notional across ten markets. In three model markets, wallets' notional-weighted ten-second markout rankings persist across adjacent ten-day windows, and on a 100-millisecond grid, eleven features describing previously top-scored wallets raise one-second out-of-sample R2 from 10.88%.

Jingqiu Ding, Daojing Zhai